Skip to main content

Full text of "Removal of selenium from water by ion exchange : review of literature and brief analysis"

See other formats

Removal of Ionic Selenium 
From Water by Ion Exchange 

Review of Literature and Brief Analysis 

August 1985 

Prepared for 

U. S. Department of the Interior 
Bureau of Reclamation 


Cal C. Herrmann 

Water Thermal and Chemical Technology Center 

University of California, Berkeley 

This report presents the results of a study undertaken 
to determine the potentiail for use of ion-exchange processes 
to remove selenium from agriculturcil drainage water. The 
study Wcis funded by the U.S. Bureau of Reclamation as part 
of the FedereLL -State Interagency San Joaquin Valley Drainage 
Program. Publication of the findings and reconmendations 
herein should not be construed as representing the concur- 
rence of either the Bureau of Reclamation or any other Fed- 
ersLL or State agency peurticipating in the Drainage Program. 
Also, mention of trade names or cammerciaLl products does not 
constitute endorsement or recommendation by the agencies. 
The purpose of this report is to provide the Drainage 
Program agencies with information and eiltematives for 
further consideration. 

The San Joaquin Vcdley Drainage Program was established in mid-1984 
and is a cooperative effort of the U.S. Bureau of Reclamation, U.S. Fish 
and Wildlife Service, U.S. GeologicaQ Survey, California Department of 
Fish and Game, and California Department of Water Resources. The 
purposes of the Program are to investigate the problems associated with 
the drainage of agricultural lands in the San Joaquin Valley and to 
develop solutions to those prctolems. Consistent with these purposes, 
Program objectives address the f clicking key areas: (1) Public hecilth, 
(2) surface- and ground-water resources, (3) agricultural productivity, 
and (4) fish cind wildlife resources. 

Inquiries concerning the San Joaquin Valley Drainage Program may be 
directed to: 

San Joaquin Veil ley Drainage Program 
Federal -State Interagency Study Team 
2800 Cottage Way, Room W-2143 
Sacramento, California 95825-1898 

Review of Literature and Brief Analysis 

Prepared for 

U.S. Department of the Interior 
Bureau of Reclanation 
Mid-Pacific Region 
2800 Cottage Way 
Sacramento, California 95825 

Cell C, Herrmann 
Water Thermal and Chemical Technology Center 
University of California, Berkeley 
47th Street and Hoffman Blvd. 
Richmond, Cailifomia 94804 

Contract No. 5-PG-20-06830 

August 1985 


Abstract. iii 

Introduction. 1 

Approaches. 1 

Digest of principal literature related to selenate. 3 

Digest of selenite and other related literature. 5 

Cost analysis. 5 

Summary and conclusions. 8 

Selenate-related references. 11 

Selenite-related references. 18 

Other selenium references. 22 
Appendix: "Selenium Removal Using Ion Exchange - Application 
to Kesterson Reservoir," Gerhard Klein, Symposium on 

Selenium in the Environment, Fresno, CA June 1985 25 


The recent literature related to ion-exchange separation and 
removal of selenium compounds is reviewed for applicability to 
agricultural drainage water improvement. One preliminary state- 
of-the-art cost estimate, assuming no exchanger selectivity but 
using an efficient regenerant, is thirty million dollars for 
treatment of the water contained in the Kesterson Reservoir to 
obtain removal of selenate along with sulfate. A byproduct 
advantage of this process is that the scaling potential of the 
water is substantially reduced, so that desalination of the water 
should require little additional pretreatment cost. A 
hydroquinone-based reducing resin currently being tested shows 
promise for continued treatment of San Luis Drain water at a cost 
of $1 0-20 per acre-foot exclusive of pre-and post- treatment. 



This review began as an attempt to identify a useful process 
for economically removing trace selenium from agricultural waste- 
waters. During its development, it broadened to include 
considerations not only of ion-exchange processes, but also of 
sorption principles as an aid to concentration or removal. 

The principal difficulty is felt to be the removal of 
selenate in the presence of large amounts of sulfate, chemically 
very similar. The portion of selenium present as selenite salts 
is a minor part of the toxic material, and is readily removed by 
known treatments, including ion-exchange. 

The possibility of reducing selenate to selenite for 
disposal, although a reaction requiring excessive activation 
energies in the laboratory, seems worthy of investigation on the 
possibility that an ion-exchange or redox resin might serve a 
catalytic function. 

References at the end of this report are divided into those 
relating to 1) selenate, 2) selenite but not selenate, and 3) 
others. Insofar as practical within the time available references 
from the first list were examined, but a few have been seen only 
as abstracts, particularly some from less common Russian and 
Japanese sources. The maximum number of possibly interesting 
selenium citations have been identified for later interest, 
though only the selenate references are discussed. Abbreviated 
abstracts are included in the citation lists, as in our database. 
Literature on related chemistry not yet applied to selenium is 
partially listed. 


Possible approaches to the problem of selective selenate 
removal are: 

1: Choice of resin or resin type, 
2: Metal-cation exchanger, 
3: Reduction of selenate. 

1 . Selective Resin Exchangers 

Clifford and Weber (1983) have recently given guidelines 
on monovalent-divalent selectivity which could be extended to 
selenate-sulf ate selectivity, since ion size is involved. They 
found that much greater selectivity for monovalent-divalent 
exchange could be obtained from epoxy, phenol-formaldehyde, and 
acrylic based amine resins than from styrene-divinylbenzene based 
amine resins. The factors responsible for this selectivity seem 

applicable not only to monovalent-divalent selectivity, but also 
(to a lesser degree) to selectivity between divalent ions of 
differing size. Whether the difference in size between sulfate 
and selenate (roughly 10%) is sufficient to achieve substantially 
enhanced ion-exchange selectivity remains to be seen. 

2. Metal-cation Exchangers 

Vernon (1982) found that copper attached to certain chelat- 
ing resins retained near -stoichiometric amounts of co-ions. 
Chloride co-ion was shown to be highly labile. For an 8-hydroxy- 
quinoline resin, sulfate co-ion was principally attached as 
RCuHSO^, while for the closer sites of a polyhydroxamic acid 
resin, sulfate co-ion was principally attached as (RCu)2S04. This 
distinction, between singly bonded and doubly bonded sulfate, 
could be expected to affect the sulfate retention differently 
than selenate retention, because of the difference in ion size. 
Thus we could hope that for a wide-spacing resin, selenate would 
be more retarded than sulfate. 

3. Reduction Processes 

Laboratory reduction processes for selenate tend to involve 
heat or strong acid, neither of which is convenient for treating 
large volumes of water. The standard reduction potential however 
is not far from the hydrogen potential, so that it should be 
attainable if the activation energy were moderated by a 
catalytic effect. Immobilization of selenate by an exchanger 
might aid the reduction either by reducing the activation energy 
(or entropy) required or by concentrating both selenate and 
sulfate temporarily from the bulk water. This provides the 
possibility, if essential, of exchanging an amount of (selenate 
plus sulfate) equivalent to a bed capacity, then subjecting the 
bed of exchanger to reduction conditions, and exchanging the 
next volume of water as a displacement of the first adsorbate 
rather than regenerating the bed to a chloride or other more 
labile form. Ames (1984), in a basalt sorption study, identifies 
hydrazine as able to reduce selenate. 

A particularly interesting approach is inclusion of the 
reducing agent in the resin: the exchanger functions then 
as a temporary retardant of anions in the water stream, long 
enough for the chemical reaction to occur: regeneration of 
exhausted resin must be by re-reduction. Reducing exchangers 
exist in two types, one with an organic reducing function 
incorporated, as hydroquinone : this is currently under 
study (Rohm&Haas Co., personal communication). This type of resin 
is not known to be in large-scale production, so may have a 
limited feasibility if the tests are successful. The other 

type, an exchanger-bound metal cation in a reduced state, may 
be more readily obtainable. One type of this resin was 
marketed as Duolite S-10, copper ions as reducing agent 
immobilized on a chelating resin. Chelating resins of the 
Dowex-IA type have remained in use. The possibility of choosing 
a suitable cation for chelation and reaction exists, and was 
noted by Helfferich (1962) in table 12-1 of his chapter on the 

Desirable criteria for such a reducing exchanger are: 

1. positive charge, as anion-attractant , 

2. 2-electron change, 

3. redox potential near that of hydrogen, 

4. moderate, but not extreme, retention of sulfate and selenate, 
supposing that they are similarly retained. 

Selenite formed, if retained strongly on the exchanger, 
could be eluted in the regeneration process, or if not retained 
strongly can be adsorbed conventionally. 

Digest of Principal Literature Related to Selenate 

Examination of recent publications indicated many studies 
done for other purposes: these are a source of valuable clues to 
selenate bonding and selection, but are not complete solutions. 
They are identified in the categories: ion exchange, solubility, 
ligands and complexes, and preparation and chemistry. 

1 . Ion Exchange 

Dreipa et al. (1980, 1981) separated selenate from tellurate 
on anion-exchangers from acid solution, eluting with strong acid 
(5% H2SO4). 

loffe et al. (1972) determined exchange constants for sele- 
nate and selenite on a strongly basic exchanger. 

Kazantzsev et al. (1972) studied similar sorptions, as 
functions of pH. 

Maneval (1984) determined selenate and selenite isotherms on 
a strong base exchanger, in comparison with sulfate, chloride and 
nitrate. No separation of selenate from sulfate was found on 
Dowex 2x8 near neutrality. 

McCarthy et al. (1983) studied selenate separation on a HPLC 
tertiary amine column without sulfate. In another study 
Williams (1983) found a greater retention time (27.2 minutes) for 
selenate than for sulfate (22.8 minutes). 

Petkova et al. (1984) separated selenate and selenite from 
copper sludge, on a strong-base exchanger in HCl. 

2. Mobility in Mineral Environments. 

Ahlrichs (1983), Ames et al. (1982, 1984) found selenate 
mobile and selenite retarded in mineral environments. 

Gary (1973) reported effects of other anions on selenium 
solubilities in soil, and Ylaranta (1982) reported on extractions 
from soils incubated with selenate and selenite. 

Presser and Barnes (1984) reported selenium speciation at 
Kesterson, and Robberecht and Van Grieken (1982) reported 
selenium determination methods. 

3. Solubility. 

Because solubility is closely related to chemical 
bonding processes, the following studies are notable. 

Bol'shakova et al. (1971) reported solubilities of selenate 
alums . 

Erdenbaeva (1975,1977) reported on solubility products and 
reduction potentials of selenium and tellurium compounds. 

Gospodinov (1984, 1984') determined solubility isotherms for 
zinc and copper selenate. 

Maneva and Stoicheva (1973) determined sodium selenate 
solubilities in organic solvents. 

Novikov and Zakrevskaya (1974) measured coprecipitation 
of selenium and tellurium species with zirconium hydroxide as 
functions of pH. 

Prumova and Selivanova (1970) compared solubilities and 
enthalpies for a series of selenate and sulfate alkali magnesium 
double salts. 

Yanitskii and Patkauskas (1970) reported solubilities in 
the sodium selenate-selenite-water system. 

4. Ligands and Complexes. 

Most selenium-related ligand studies are of divalent 
selenide ligands. A few remain of interest. 

Marov et al. (1981) have information on selenate-der i ved 
organic copper complexes. 

Neogy and Nandi (1982) have made magnetic measurements of 
ligand field properties on ytterbium selenate. 

Richardson and Hilmes (1975) analyzed the structure of zinc 
selenate from the crystal field point of view, and found the 
structure similar to that of zinc sulfate. 

Shukla and Pandey (1978) used an EPR method on cobalt 
potassium selenate, and Waplak et al. (1975) used EPR on sodium 
selenate for structural analysis. 

4. Preparation and Chemistry 

Blanka (1974) used ion exchange in synthesis of selenic 
acid . 

Chagas (1979) made triangular-sweep electrochemical measur- 
ements on a platinum rotating-disc electrode in the presence of 
partial coverages by selenate and selenite. Oxidation, but not 
reduction, selenium peaks were visible; lowered platinum oxida- 
tion and reduction currents appeared as a function of coverage. 

Digest of Principal Literature on Selenite and "Other" References 

Included principally for future use of those concerned with 
related selenium separation and chemistry are the items found 
concerning selenite as well as those items for which valence 
state was not readily available because of incomplete abstracts 
or because the documents were not readily available for 
examination. References include those on use of selenites as ion- 
exchange materials themselves, since these may be of value as 
indications of chemical bonding and bonding mechanisms between 
selenite and other materials. 

Cost Analysis 

An analysis of selenate removal by ion exchange, from 
Kesterson water, has recently been made by Klein (1985). The 
analysis initially evaluates "ideally selective" exchanger costs, 
then, as a more realistic state-of-the-art condition, assumes an 
unselective strong-base exchanger, with purchased calcium 
chloride as regenerant. This is presented as a worst case, to be 
improved upon if possible by provision of more selective 
exchangers. Based on total anion exchange of 10 billion liters of 
(0.15 Normal) Kesterson water, a resin cost of $270,000 was 
calculated, to yield a total plant and process cost of just under 
$3 million, exclusive of regenerant. Klein proposed calcium 
chloride as a readily available, moderate cost chemical (in the 
quality used for ice removal on roads). Used as a precipitating 
regeneration, as practiced in the Los Bancs and Firebaugh 
desalination test stations (but there as sulfate regenerating a 
calcium-loaded cation exchanger), a solid, readily disposable 
waste is produced. The stoichiometric regenerant cost was 
calculated as $21 million. A total cleanup cost of $30 million is 
therefore suggested, neglecting any salvage value of the waste 
(calcium sulfate containing less than 0.1% calcium selenate 
contaminant). The resulting water free of sulfate as well as 
selenate has low scaling potential, and could be used as feed for 
reverse-osmosis desalination for reuse, with little or no 
pretreatment . 

It is interesting to note, for comparison of costs, that 
Merriam (1985) has concluded that the Kesterson water could be 
diluted to an environmentally satisfactory selenium level, using 
irrigation-quality water costing $50 /acre-foot , for about the 
same price ($32 million, for a 75-fold dilution). 

As a single-time cleanup of a contamination which ha's 
accumulated over years, the above cost level (more than $2000 per 
acre-foot of treated water) is tolerable, but for continuing use 
on the agricultural drainage water of the San Luis Drain, this 
cost is high. 

In the paper cited above, Klein also evaluates costs for the 
"ideally selective exchanger." To treat ten billion liters of 
Kesterson Reservoir water in 1000 days, that is ten million 
liters/day, a tiny resin cost (at the manufactured cost of 
conventional resins) of $4.46 was found. It was concluded that 
the exchanger cost was insignificant compared with process costs 
for pumping and regeneration. 

Recently the hydroquinone reducing resin, mentioned above 
under the Reduction Processes heading, has been shown to approach 
the "ideally selective exchanger" performance (Murphy 1-985). 
While not an ion-exchange process, it may be evaluated similarly. 

For practical convenience the above treatment rate, 10 
million liters/day can be approximated as 10 acre-feet/day. The 
San Luis Drain water is reported at more than three times the 
assumed Kesterson selenium concentration. Since some organically- 
bound selenium is present, which must be removed by other (e.g. 
adsorption) processes, 0.3 ppM will be chosen as the average 
selenium level removable by this process. Cost for this specialty 
resin would be the minimum-batch cost, or $300/cu.ft. times 300 
cu. ft., or $100k. The above treatment rate, with four 
regenerations /day , would require less than one cubic foot of 
resin. Therefore the minimum batch would allow for 300 resin 
replacements, on the expectation that the lifetime of a reduction 
resin is limited. As a guess, resin life from ten to 100 cycles 
could be conservatively supposed. The long-term resin cost then 
might be one to ten dol lars /acre- foot ; less if the resin proves 
more durable. 

The practical success of an exchanger as ideally selective 
requires that there are no significant interferences, that is in 
the case of a reducing resin, that there are no other substances 
in the feed water that can be reduced. Water constituents that 
might react with the exchanger are oxygen and ferric iron. 
Insofar as resin capacity is taken by such species competing with 
selenium, the amount of resin needed per acre-foot per cycle 
would be increased, and the regenerant need would also increase. 
(Such interfering substances might be chemically reduced or 
removed before the feed is passed through the resin bed, 
however . ) 

Regenerant cost, assuming a calculated equivalence of about 
0.3 pounds of sodium sulfite per acre-foot, would be less than 
$1 /acre-foot unless extreme interferences are present. Removal of 
selenium from the resin bed is expected to be only an occasional 
process, so would not contribute substantially to the overall 
process cost. Pumping cost, calculated as 1.4 horsepower-hour per 
foot of head per acre-foot of water, is about $1 /acre-foot for an 
estimated 10 foot head required. 

Considering the number of unknown factors, practical 
selectivity, resin life, plant amortization, supervision and 
maintenance costs, only a crude estimate can be made. The 
principal operating costs as itemized above can be accommodated 
by a cost prediction of $10-20 per acre-foot treated. This is 
sufficiently moderate that investigation of the unknown factors 
is justifiable for possible application in conjunction with, or 
without, reverse osmosis desalination for reuse of the drainage 
water . 

Still more recently a preliminary cost estimate has been 
made for the hydroquinone reducing exchanger process (Bureau of 
Reclamation internal report, Denver, May 15, 1985). This 
estimate, based in part on an EPA drinking water cost study (EPA 
1979), is much more comprehensive than our estimate above, based 
only on resin and regenerant cost expectations. It is interesting 
to make certain comparisons. The complete process cost (for 
"Concept B" in the report) is $337/acre-f oot of water treated. 

The cost per acre-foot of water treated, for the major 
components (calculated from tables B-1 and B-4 of the report) are 
as follows: 


Dollars per 

Main elements 


I . Chemicals 

G. Capital/ 


J .Labor 






Resin 37. 

Sand filter, polish. 1 6 . 4 
" , raw water 10.9 
Backwasher 8.5 

Regenerant sump 6,6 
Backwash sump 5.5 


K. Energy 

1 2.7 

H. Maintenance 


The first observation is that this projection assumes a 
monthly resin regeneration, rather than the more common choice of 
4 regenerations per day, assumed above. This may be required by 
some process condition that is not obvious, however this in large 
part accounts for the greater resin and associated facility costs 
found by this report. 

A second observation is that labor cost seems based on 
allocation of a minimum staffing average of 2 persons on duty, 
as 24-hour supervision. 

The first conclusion is that the principal cost elements are 
not those of the resin process itself, but the pre- and post- 
treatment combined elements. 

A second conclusion stems from the further observation that 
the complete process cost cited above is a major portion of 
typical cost estimates for desalination-f or-reuse processes. It 
is unclear whether it practically follows that this selenium 
removal process could optimally be combined with a desalination 
facility, thus sharing much of the pre- and post- treatment, 
labor and plant operating costs; or whether this selenium removal 
process is effectively in competition with a desalination-f or- 
reuse water treatment process. 

As a recommendation, a further evaluation of the process by 
an experienced consulting-engineering firm seems justified, to 
more accurately estimate the many plant costs beyond the scope of 
concept evaluation, and also that this evaluation may include 
determination of the cost elements combinable with a Los Banos- 
type desalination plant. 

Summary and Conclusions 

The preponderant ionic species of selenium in the 
Kesterson Reservoir, and San Luis Drain feed water, is selenate. 
Previous research by the University of California Water Thermal 
and Chemical Technology Center has shown that conventional 
strong-base ion exchangers cannot discriminate between selenate 
and sulfate ion, which is the principal anionic species in this 
water. Other exchangers, e.g. activated alumina, have been 
reported selective for selenite removal. Therefore, future 
investigations could include study of: 

1. An overall theoretical assessment of the potential role of ion 
exchange in selenium removal, that would take into account the 
mul t icomponent anionic composition of the drainage water, and 
acceptable regeneration methods. Such an assessment would be 
based on the assumption that an exchanger of any desired 
selectivity could be obtained or provided. Study of the problem 
to this level could provide some basis for a semiquantitative 
economic assessment. 

2. A search for, and possible synthesis of, ion exchangers 
exhibiting a significant preference for selenate over sulfate 
ion, and capable of being regenerated economically and in an 
environmentally acceptable way. 

3. Reduction of selenate to selenite or to selenium. Since 
selective removal of selenite is more available than selective 
removal of selenate, reduction of selenate to selenite can be 
examined a) by reducing sorbants, or b) by chemical reduction 
while bound to a nonselective exchanger. Process evaluation 
requires knowledge of possible interferences (oxidants) in the 
chosen feed water, and knowledge of resin life. 

4. A study of precipitation processes, to determine the extent to 
which selenate is co-precipitated with sulfate in a calcium 
chloride regeneration of an anion-exchange resin. 


The Principal Investigatorship of Professor A.D.K. Laird 
and stimulating discussions with members of the Ion Exchange 
group of the University of California's Water Thermal and 
Chemical Technology Center are gratefully acknowledged. 

This study has been made with the financial support of 
the U. S. Bureau of Reclamation. In this connection, the author 
wishes to thank Dr. Edwin Lee of that organization for his 
understanding and encouragement. 


Selenate-related References 

Ahlrichs, J. S. /I 983. Movement of selenite and selenate by 
saturated and unsaturated flow in east Texas overburden 
Texas A&M, Univ. Microfilms DA8329893. 

20-60m cores of primary lignite outcrop, pH's 3,5,7,9. Soil TLC. 
Greatest Se(IV) sorption pH3, near pK1=2.75. (497 of 500 ug/g). 
Less than 1.2 ug/g retained at pH 2&9. (pK2=8.5). Se(VI) moved 
with front: Rf=0.96 at pH 7,9. At pH 3 Rf=0.76. concl: Se(VI) 
non-specif ically adsorbed (pH<7), Se(IV) both ppt. and ligand- 
exchange mechanisms 

Ames, L. L.; P. F. Salter; J. E. McGarrah and B. A. Walker /I 982 

Selenium sorption on a Columbia River basalt 

PNL-SA-10750 (report: NTIS) 

Na2Se04 + water + hydrazine + basalt, 40-60C. Co-Ct=Kt(exp b): Co 

initial Se04, Ct cone, at time t, K,b determined. Desorption 

(60C) C=At(exp d). Freundlich isotherm. 

Ames, L. L.; P. F. Salter; J. E. McGarrah and B. A. Walker /I 984 

Selenate-selenium sorption on a Columbia River basalt 

Chem. Geol. 43,287 

Na2Se04 + water + hydrazine (reducing conditions, Se4 vs Se6) + 

basalt, 40-60C. Co-Ct=Kt(exp b): Co initial Se04, Ct cone, at 

time t, K,b determined. Desorption (60C) C=At(exp d). Freundlich 


Blanka, B. /1974. Preparation of selenic acid by ion exchanger. 

Chem. Zvesti 28, 298 

K, Mn selenates. From H2Se03+KMn04 

Bol'shakova, N. K.; E. A. Zalogina; N. M. Selivanova /I 971 

Solubility and thermal decomposition of potassium, rubidium, and 

cesium aluminum selenate alums 

Zh. ^Neorg. Khim. 16, 378 /Russ. JIC 16, 197 

heats of solution 

Gary, E. E. and G. Gisse 1-Nielsen /1973. Effect of fertilizer 
anions on the solubility of native and applied selenium in soil 
;Soil Sci. Soc. Amer. Proc. 37, 590 
phosphate, nitrate, sulfate 

Chagas, Helio C. /1979. Isopotential points in the 
electrosorption of selenite, selenate, selenide, and tellurite at 
the platinum rotating disk electrode 
Can. J. Chem. 57, 2560 
in sulfuric acid 


Clifford, D. and W, J. Weber Jr. /1983. The determinants of 

divalent/monovalent selectivity in anion exchangers 

React. Polymers 1, 77 

Charge separation, resin type determine N03, Cl vs S04 

selectivity. DVB gives S04/N03 ratio 1.7, epoxyamine gives 137. 

Isotherms, titration curves. 

Dreipa, E. F.; V. S. Pakholkov and S. A. Luk ' yanov /1981 

Sorption of tellurium ions from aqueous solutions by anion 

exchangers and ampholytes 

Zh. Prikl, Khim. (Leningrad) 54, 1040 

Sepn. H2Se04 from H6Te06 on EDP-IOp. Functional groups of several 

resins, comparisons. 5% H2S04 elutant. IR spectra. 

Dreipa, E. F.; V. S. Pakholkov; S. A. Luk'yanov /I 980. Sorption 
of selenium(IV) and selenium(VI) from aqueous solutions of its 
acids by anion exchangers and ampholytes 
Zh. Prikl. Khim. (Leningrad) 53,54 

Soln. in 0.1 N H2S04 eluted with 5%H2S04. Se(IV), Se(VI) sepn. on 
EDE-lOp resin. Other resins tested: AV-17x8, AN-31, AN-2F, AV- 
18x8, AV alpha-8p, AV- beta-8p, AV- gamma-8p, AVKh-8p. 
Aminophosphoric acid ampholytes: ANKF-2G, -3G...cap Se(VI)>Se(IV) 

Erdenbaeva, M. I. /1975. Solubility product of selenium and 

tellurium compounds 

Vestn. Akad. Nauk Kaz. SSR 1975^2, 51 

-ide, -ite, -ates. solubility, reduction. 

Erdenbaeva, M, I. /I 977. Solubility products of some selenium and 

tellurium compounds 

VINITI 3215-77 

copper, mercury chalcogenates , -ites , reduction potentials 

Flint, C. D. and M. Goodgame /I 967. Magnetic and spectral studies 

including metal-ligand vibrations of thiourea complexes of some 

salts of divalent metals with oxyanions 

J. Chem. Soc. A 11, 1718 

Zn, Cd, selenate, Co2, thiosulfate, Ni2 complexes. X-ray 

diffraction, magnetic moments. IR spectra. 

Ghosh, U. S.; R. N. Bagchi ; A. K. Arun; S. N. Mitra /1967 

Orthorhombic ligand field theory of the magnetic behavior of 


Indian J. Phys. 41 , 286 

Crystal field, molecular orbitals, magnetic susceptibility 


Gospodinov, G. /I 984. Solubility diagram in the zinc selenate- 

hydrogen selenate- water system and some properties of the 

compounds obtained 

Z. Anorg. Allg. Chem. 517, 223 

Hydrates, nH20= 0,1,5,6. Liquid phase to 40-60% H20. ZnSe03 + 

H202 -> ZnSe04.+ (590C)-> ZnSeOS. 

Gospodinov, G. /1984'. Solubility isotherms in the copper(II) 

selenate-selenic acid-water system at 25 and 100 degrees C: the 

solubility polytherm in the copper(II) selenate-water system, and 

some properties of the compounds obtained 

Z. Anorg. Allg. Chem. 513, 213 

Ternary and thermal derivative diagrams: mono, tri, penta 

hydrates. Thermal reduction (490 C) mechanism. 

Hoover, T. B. and G. D. Yager /I 984. Determination of trace 

anions in water by multidimensional ion chromatography 

Anal Chem 56, 221 

on Dionex, collected and rerun Se(VI) shoulder on S04 

loffe, V. P.; I. N. Baklashova; M. N. Yakimova; G. M. Kolosova; 
R. N. Rubinshtein /I 972. Determination of exchange constants for 
anions of dibasic acids on strongly basic anion exchangers 
Zh. Fiz. Khim. 46, 2094 /Russ. JPC 46, 1195 
AV-17, calculations, values (partial). 

Kazantzsev, E. I.; V. A. Prokhorov; M. K. Makarov /I 972. Sorption 

and separation of ions of vanadium(V), s e 1 e n i u m ( V I ) , 

molybdenum(VI) , and rhenium(VII) from nitrate solutions using an 

ampholyte and AV-17 anion exchanger 

Izv. Vyssh. Ucheb. Zaved. Tsvet. Met. 15, 87 

Function of pH . 

Klein, Gerhard /I 985. Selenium removal using ion exchange 
Proceedings of the Symposium on Selenium in the Environment, 
California State University, Fresno, June 1985. 

Assumes no selectivity on strong-base resin, regeneration to 
calcium selenate, sulfate waste for solid disposal. Cost analysis. 

Maneva, M. and M. Stoicheva /I 973. Solubility of sodium selenate 

and the behavior of sodium selenate decahydrate in organic 


Monatsh. Chem. 104, 356 

dehydration of decahydrates in MeOH, EtOH, partial in acetone. 

Solubilities 18-974 mg/1 


Maneval, James /I 984. Selenium removal from drinking water by ion 


Thesis, University of California, Berkeley 

Se(IV), Se(VI) on strong-base ix. isotherms. Fe3 on weak acid 

exchanger holds Se(IV). no Se(VI), sulfate sepn on resin. 

Marov, I. N.; V. K. Belyaeva; E. Hoyer; VI. Dietzsch; R. Kirmse 

/1981. Formation of mixed sulfur- and selenium-containing 

copper(II) complexes 

Koord. Khim. 7,1471 

Ligands: isomaleonitriled selenate, and thiols. ESR parameters -> 

Ke for CuL2 + CuL2' -> CuLL'. Ring size (4 or 5) 

McCarthy, J. P.; J. A. Caruso and F. L. Fricke /I 983 

Speciation of arsenic and selenium via anion-exchange HPLC with 

sequential plasma emission detection 

J. Chromatogr. Sci. 21, 389 

Strong AEX: Nucleosi 1-NH ( CH3 ) 2 on silica support, elution by 

ammonium acetate - dihydrogen phosphate pH 4.6 -> 6.9. No sulfate 

included. Elution order As3-Se ( IV ) -As 4-Se (VI). Detection limits 

3.9 ng/sec Se = 140 ng Se(IV), 91 ng Se(VI). 

Merriam, Marshal F. /I 985. (personal communication) 

Mignonsin, E. P. /1974. Neutron radioact i vat ion analysis of pure 
selenium. Application of resin ion-exchange chromatography to the 
systematic analysis of selenium 
J. Radioanal. Chem. 19, 33 

Anal, for impurities in Se. Dowex 1x8. Se dissolved in cone. 
HN03, evap., dissolved in HF-HN03 dil. to M, NH3 to pH 5-6. Se 
retained on elution w. ION HCl but elutes w. 6 N HCl, H20. S 
elutes w. NH4C1. 

Murphy, A. (1985) Internal report, Bureau of Reclamation 
Eng/Research Center, Denver, Colorado. 

Neogy, D. and J. Nandi /I 982. Magnetic measurements on ytterbium 

selenate octahydrate and the nature of its ligand field 

J. Chem. Phys., 76, 2591 

Yb2(Se04)3.8H20. 90-300 K. Crystal field: tetragonal symmetry RE 

sulfates & selenates. 

Novikov, A. I. and T. M. Zakrevskaya /1974. Coprecipi ta t ion of 
arsenic(V) and arsenic( 1 1 1 ) , selenium(IV) and seleni urn ( VI ) , 
tellurium(IV) and tellurium( VI ) with zirconium hydroxide 
Radiokhimiya 16, 769 
valence detn. sorption, prep., properties. Functions of pH . 


Petkova, E.; Kh. Vasilev and E. Dekova /I 984. Ion-exchange 
sorption of selenium(VI) and selenium(IV) on Wofatit SBW anion 
exchanger using the exhaustion-curve calculation method 
Metalurgiya (Sofia) 39,21 

Cu refining sludge melted with Na2C03 and dissolved in HCl. Ions 
extracted on resin-Cl and eluted with HCl -> 50 g/dm3 Se. Calcn. 
method of Gromoglasov & Kalpakchiev. 

Presser, T. S. and I. Barnes /1984. Selenium concentrations in 

waters tributary to and in the vicinity of the Kesterson National 

Wildlife Refuge, Fresno and Merced counties, California 

uses Water Investigations Report 84-4122 

Laboratory methods reviewed and described; map of sampling points 

(Kesterson and source areas); results: total and selenite, 

sulfate, sodium, dOI 8 & d-D H20, discussion of organics. 

Prumova, L. /I 975. Phy sicochemical study of double salts of 

M2IMII(X04)2.6H20 type 

Khim. Ind. (Sofia) 47, 452 

alkali, alkaline earth double salts with selenate, sulfate. 

Solubilities, properties. 

Prumova, L. A. and N. M. Selivanova /1970. Solubility in water 

and heats of dissolution of binary magnesium sulfate and selenate 


Tr. Mosk. Khi.- Tekhnol. Inst. 67, 15 

Double salts M 2 Mg ( Se04 ) 2. 6H20 & S cmpd. M=K,Rb,Cs. deltaH 

(kcal/mole) for K,S: 9.36, Se: 7.49; Rb,S: 11.22, Se: 10.33; 

Cs,S: 10.55, Se: 8.98 +/- 0.1 about. 

Richardson, F. S. and G. Hilmes /I 975. Theory of natural optical 

activity in crystalline copper(2+)- doped zinc selenate 


Mol. Phys. 30, 237 

dipole and quadrapole moments, crystal field, ligand field, 

circular dichroism, vibronic coupling. H20's coordinated with 

copper. Structure as NiS04,6H20 

Robberecht, H.; and R. Van Grieken /1982. Selenium in 

environmental waters: determination, speciation and concentration 

levels . 

Talanta 29, 823 

analyt. methods reviewed: spectre, turbidity, GLC, HPLC, Echem, 

fluorometry, neutron activ. Detn. limits. 


Salter, P. F. and G. K. Jacobs /I 982. Evaluation of radionuclide 

transport: effect of radionuclide sorption and solubility 

Sci. Basis Nucl. Waste Manage. 11, 801 /Mater. Res. See. Symp. 

Proc, W. Lutz, ed., Elsevier. 

Travel times, oxidizing vs reducing conditions (2E5 vs 7E7 

yrs/meter at low water flow) 

Sarquis, M. and C. D. Mickey /1980. Selenium part 1: its 
chemistry and occurrence 
J. Chem Ed 57, 886 
Review; part 2 not found. 

Shukla, S. R. and S. D. Pandey /I 978. EPR of manganese( 2 + ) in 
cobalt potassium selenate hexahycrate Tutton salt 
Phys. Status Solidi B 85, K1 03 
crystal field, ligand field 

Staub, C; J. Buffle and W. Haerdi /1984. Measurement of 
complexation properties of metal ions in natural conditions by 
ultrafiltration: influence of various factors on the retention of 
metals and ligands by neutral and negatively charged membranes 
Anal. Chem. 56, 2843: see also ibid p. 2837 (method) 
Similar (95%) retention of sulfate and selenate by Amicon UM-05 
(negatively charged) but selenate less retained (80%) than 
sulfate (95%) by Nucleopore MWCO-C-500 (neutral). 

Sorg, T. J. and G. S. Logsdon /I 978. Treatment technology to meet 

the interim primary drinking water regulations for inorganics: 

part 2 

J. AWWA (:uly) 379 

coagulation, lime results: As ana Se. 

Uchida, H., Y. Shimoishi and K. Toei/1980. Gas chromatographic 
determination of selenium (-II, C, IV, and VI) in natural v/aters 
Envir. Sci. and Technol. 14, 541 

Se 6 reduced to Se 4 with HBr, reacted with 1 , 2diamino-3 , 5dibromo 
benzene, extracted into toluene, gc gives 2 ng Se/1 of water. 

Vernon, F. /1982. The role of the co-ion in chelating ion 

exchange processes 

React. Polymers 1, 51 

Resins which give RCuCl vs R2Ci ; Resins which give RCuH; 04 vs 

(RCi!)2S()4. CI found labile. 

Waplak, S.; A. Malecka; „' . Sl:ank)wski; L. A. Shuvalov /I 97 3. EPR 
study of chromium(3+) doped sodiu.n trihydroselenate monocry tal 
Acta Phys. Pol. A47, 809 
Electron spin resonance, crystal and ligand field theories 

1 6 

Williams, R. J. /1983. Determination of inorganic anions by ion 

chromatography with ultraviolet absorbance detection 

Anal Chem 55, 851 

retention times on Dionex "anion" column 

Yanitskii, I. V. and R. Patkauskas /1970. Solubility and 
crystallization in a sodium selenate-sodium selenite-water system 
Zh. Prikl. Khim. (Leningrad) 43, 522 

phase, chemical equilibria. Solubilities as functions of 
composition, 0-60 C. 

Ylaranta, Toiro /I 982. Sorption of selenite and selenate in the 


Ann. Agric. Fenn. 22, 29 

1-3 mo. incubation of Na selenate and selenite, and extraction 

with hot water gives: (clay) 0-4%; (fine sandy soil) 1-9%; (peat) 

21-41%. Liming increases extraction. Other extractants: NH40H, 

NH4C204, KH2P04. -.0001 M. 

Yu, M., G. Liu and Q. Jin /I 983. Determination of trace arsenic, 

antimony, selenium and tellurium in various oxidation states in 

water by hydride generation and a t o m i c - a b s o r p t i o n 

spectrophotometry after enrichment and separation with thiol 


Talanta 30, 265 

cotton+thioglycollic acid holds Se(IV) to pH3, Se(VI) beginning 

4N HCl. Se(VI) reduced to Se(IV) with 0.1% TiCl3 


Selenite References 

Andrews, R. W. and D. C. Johnson /1976. Determination of 
selenium( IV) by anodic stripping voltammetry in flow system with 
ion exchange separation 
Anal. Chem. 48, 1056 
cation-exchange chromatography 

Barkovskii, V. F. and Yu. A. Pavlovich /1977. Sorption of 
selenium and tellurium on molybdophosphor ic acid salts under 
equilibrium conditions 
Zh. Anal. Khim. 32, 1913 

Chikuma, M. ; M. Nakayama;,H. Tanaka; K. Tanaka ; T. Tanaka /I 981 

Preparation of the resins for the collection of pollutants by the 

conversion of the ion-exchange resin with some terf unct ional 

reagents: application to the collection of mercuric, selenite, 

and fluoride ions. 

J. Pharmacobio-Dyn. 4,5 

alizarin, lanthanum complexes. Useful: ( 4 - sul f ophenyl )diazene- 

carbothioic acid ( 2-sulf ophenyl ) hydrazide, di Na salt 

DeCarlo, E. H. and H. Zei 1 1 in/ 1 98 1 . Simultaneous separation of 

trace levels of germanium, Antimony, Arsenic, and Selenium from 

an acid matrix by adsorbing colloid flotation 

Anal Chem. 53, 1104 

Fe(III) carrier, pH 4-5.5. Sulfate is present, separation not 

discussed . 

Ebert, M.; Z. Micka and I. Pekova /1982. Synthesis of nickel 

selenites, their solubility and bonding in them 

Coll. Czech. Chem. Commun. 47,2069 

Ni(HSe03)2.2H20; NiSe03.2H20. IR reflectance spectra, magnetic. 

Se-0 force constants. 

Ebert, M.; Z. Micka and M. Uchy t i lova/ 1 98 4. Zinc selenites: the 
solubility diagram and its use for the isolation of the 
compounds, their spectral features and thermal behavior and the 
strength of bonds in them 
Coll. Czech. Chem. Comm. 49, 1653 

Coll. Czech. Chem. Comm. 49, 1653 

phase diagrams 0, 25C; prep. Z n ( HSeO 3 ) 2 . 2 H 20 , ZnSe205 

spectra, force constants. 


Eggers, H. and H. A. Ruessel /I 984. Chromatography of xanthate 

complexes of As, Sb, Bi, Se, Te, and Ni. 

Fresenius' Z. Anal. Chem. 318, 278 

TLC on silica gel, A1203, cellulose, polyamide; HPLC on LiChrosorb 



Faix, W. G.; R. Caletka and V. Krivan /I 981. Element distribution 

coefficients for hydrofluoric acid/nitric acid solutions and the 

anion exchange resin Dowex 1x8 

Anal. Chem. 53,1719 

log D = 0.3 in pure HF , near zero in mixtures 

Husain, S. W. /1972. Chromatographic separation of metal ions on 

ion-exchange papers 

Analusis 1 , 314 

Ti tungstate paper. Pd, Pt, Se, Hg, Cd, alkalines, rare earths, 

transitions . 

Husain, S. W. ; M. Ghannadi-Marageh and S. Rasheedzad /I 984 
Synthesis and ion-exchange properties of cerium(IV) selenite 
J. Radioanal. Nucl. Chem. 84, 239 
Distribution coeffs. from 4 (Ni2) to 120 (Ba2): ml/g 

Idriss, K. A.; M. M. Seleim and M. S. Abu-Bakr /1980. An 
analytical study of mixed-ligand selenium(IV) complexes. The 
ternarycomplexof selenium{IV) w i thi , 1 Ophenanthrolineand eosin 
Mikrochim. Acta 2,179 
1-2-2 complex of Se-E-P at pH 3.5-4 follows Beer's law. 

Idriss, K. A.; M. M. Seleim; M. S. Abu-Bakr; M. S. Saleh /1982 
Physicochemical study of mixed-ligand selenium{IV) complexes: 
ternary complex of selenium(IV) with alizarin maroon and eosin 
Analyst (London) 107,12 

Mixed-ligand complex spectrophotometric absorption max. 560 nm, 
Beer's law 0.16-2 ug/ml. bump on AZM-EOS curve: peak shift -80nm. 
EDTA no competition. pH=7. titration: complete above pH 5.3, 
stable above. Room T formation. 

loffe, V. P. and R. N. Rubinshtein /1971. Selection of optimum 
conditions for the ion-exchange isolation of trace amounts of 
sulfuric acid from selenious acid solutions 

Sb. Nauch. Tr., Gos. Nauch. -Issled. Inst. Tsvet. Metal 34, 75 
anion exchange on AV-17 in OH form, from H2Se03 (.5N), elute with 
HCl. H2S04 enriched in eluate at low, high H2Se03 concentrations. 
Distribution coeffs. equal on exchanger in CI form in HCl. 

Kawamoto, M. and H. Murakami /1977. Recovery of selenium oxy- 

acids in wastewater 

Japan Kokai Tokkyo Koho JP 78108898 

selenous acid recovery by ion exchange 


Kock, W. and J. Korkisch /I 973. Anion-exchange separation of 
elements extractable with tributyl phosphate. V. Possibilities 
for separations in tributyl phosphate- hydrochloric acid or 
tributyl phosphate- methylene glycol- hydrochloric acid- 
containing solvent systems 
Mikrochim. Acta 1973, 245 
Weak absorption Se4, strong Te4. 

Kuroda, R.; K. Oguma; M. Suzuki; Y. Fuchu /I 975. Adsorption 
behavior of metals on TEAE-cellulose from hydrochloric acid and 
hydrochloric acid- thiocyanate media 
Bunseki Kagaku 24, 1 

Thiocyanate eluent. As, W, V, Mo, Zn, Pd, Cu, Au, Se, on anion- 
exchange TEAE. Distribution coefficients decrease with HCl 
increase, 0.001 -6N 

Liu, X.; M. Liu and Z. Hu /1983. HPIEC separation of selenium and 

tellurium and their determination 

Gaodeng Xuexiao Huaxue Xuebao 4,640 

Selenite on YSG-R4NC1 resin eluted with NaCl, NaBr or Na tartrate. 

Massee., R. ; F. J. M. J. Maessen and J. J. M. De Goeij /1981 

Losses of silver, arsenic, cadmium, selenium and zinc traces from 

distilled water and artificial seawater by so Goeij /I 981 

Losses of silver, arsenic, cadmium, selenium and zinc traces from 

distilled water and artificial seawater by sorption on various 

container surfaces 

Anal. Chim. Acta 127, 181 

tracer (75Se) method, losses insignificant for Se. 

Nabi, S. A.; A. R. Ahsan; R. A. Khan Rao /1981. Synthesis, ion 

exchange properties and applications of thermally stable stannic 

se lenophospha te : comparison with other tin(IV) based ion 


J. Liq. Chromatogr. 4, 1225 

prep., separations, comparisons with other exchangers. 

Pakholkov, V. S.; Rashchupkin, G. V.; Dreipa, E. F. /I 981 

Sorption of selenium(IV) ions by anion exchangers and zirconium 


Kompleksn. Ispol'z. Miner. Syr'ya #12,72 ■ 

Exchangers AN-31 > AN-2F > EDF- 1 OD > AV- 1 7x8 > Zr hydroxide; m H2Se03 

solns. Adsorbed HSe03 IR spectra. 

Petkova, E.; Kh. Vasilev and Z. Kalpakchiev /I 974. Use of some 

ion exchangers for the production of pure selenium 

Metalurgiya (Sofia) 5, 12 

At pH 2, to IRA 93 (01 form), eluted with 3NHC1. Better than 

strong base for removing impurities. 


Rawat, J. P. and K. P. S. Muktawat /1981. Synthesis and ion- 
exchange properties of tantalum selenite and its use for the 
separation of metal ions by ion-exchange column chromatography 
J. Liq. Chromatogr. 4,85 

Rawat, J. P. and R. A. Khan /I 981. Synthesis and ion-exchange 
properties of niobium selenite: separation of lanthanum-thorium, 
thorium-copper, lead-cadmium, zinc-lead, and lanthanum-cerium 
Ind. J. Chem. 20A,754 

weak cation-exchanger, mono acid form, capacity - 1 meq/g. OK to 
200C, and in 2N nitric acid. 

Sharma, S. D. and B. C. Lathe /I 984. Synthesis and ion-exchange 

properties of thermally stable stannic silicon selenite: 

separation of copper(2+) from zinc(2+) and manganese( 2+ ) 

J. Indian Chem. Soc. 61, 375 

prep, from Si02xH20 in NaOH + SnCl4 + Na2Se03 solutions. 

Distribution coeffs. from 1 (Mn2) to 300 (Cu2) in H20, ).2 (Mn2) 

to 200 (Cu2) in 10-2 M HN03. Hg2, Pb2 also high, Fe3 low. 

Shibata, Y.; M. Morita and K. Fuwa /I 984. Determination of 

selenium by liquid chromatography with spectrof luorometr ic 


Anal. Chem. 56,1527 

derivative detection limit 130 fg Se, at S/N=2. But blanks run 3 pg, 

Slavcheva, Yu.; E. Papova and G. Gospodinov /I 984. Solubility and 

solubility product of the selenites of the group IV elements 

Z. Chem. 24,105 

note. Solubility products in HN03, HCl, H2S04 solutions at 20C: 

Ge(Se04)2: 3.38E-26, SnSe04: 2.30E-31, PbSe04: 3.12E-13 

Tanaka, H,; M. Chikuma and M. Nakayama /1982. Some novel 
functional resins for the collection of water pollutants 
Pergamon Ser. Environ. Sci. 7, 381 
IRA-400 anion-exchanger + Bismuthiol II, for determination of Se4 

Tanaka, H.; M. Nakayama; M. Chikuma; T. Takana; K. Itoh and H. 

Sakurai /1984. Selective collection of selenium(IV) from 

environmental water by f unctionalized ion-exchange resin 

Stud. Environ. Sci. 23, 365 

new functional resin bismuthiol-II 

Zapat, W.; S. Waplak; J. Stankowski; L. A. Shuvalov /1978. EPR 

investigation of potassium trihalogen selenite doped with 

chromium(3+) ions 

J. Phys. Soc. Jpn. 44, 1600 

crystal field, f erroelasticity , ligand-field theory 


other Selenium References, including unidentified valence 

Babayan, G. G.; E. E. Kapantsyan and E. N. Oganesyaan /I 972. Ion 
exchange chromatographic separation of selenium, tellurium, and 
bismuth from sulfuric acid solutions 
Uch. Zap., Erevan. Univ., Estestv. Nauki 1972, 46 

Bahnick, D. A.; T. P. Markee; C. A. Anderson; R. K. Roubal /I 978 
Chemical loadings to southwestern Lake Superior from red clay 
erosion and resuspension /I 978 
J. Great Lakes Res. 4, 186 
clay, erosion, resuspension 

Bernier, W. E. and G. E. Janauer /I 976. In situ reduction of ion 
exchange resins as a method for preconcentration of selenium and 
other heavy metals from aqueous solutions 
Trace subst. Environ. Health 10, 323 
to elements: Se, As, Sb, Hi 

Bernotas, V. and I. A. Grachev /I 977. Development of methods for 
determining the radioisotopic purity of ruthenium-1 06 preparations 
Khim. Tekhnol. Izot. Mechenykh Soeden 1, 12 
Se sepn., ion exchange chromatography 

Bonomo, R. P.; E. Rizzarelli; S. Sammartano; F. Riggi /I 980 

Copper(II) simple and mixed complexes containing tridentate 

ligands with oxygen, sulfur or selenium donor atoms in aqueous 

solution. Spectroscopic investigations and thermodynamic 


Inorg. Chim. Acta 43, 11 

Cu selenodiacetate complex, pseudo octrahedral geometry. ESR, UV, 

enthalpy . 

Chen, Sizhen /1982. Separation of selenium and tellurium by ion 

exchange and their determination by induced reaction 

Fenxi Huaxue 10, 342 

Se eluted with 0.1N NaOH, from Zerolit 225 (low cross- 1 inking ) . 

Analysis: Au reduction induced. 

Fuller, W. H.; C. McCarthy; B. A. Alesii; E. Niebla /1976. Liners 
for disposal sites to retard migration of pollutants 
EPA-600/9-76-01 5 

Se migration, sorption by agricultural limestone, organic liners, 
ferrous sulfate, nutshells 


Geering, H. R.; E. E. Gary; L. H. P. Jones; W. H. Allaway /I 968 

Solubility and redox criteria for the possible forms of selenium 

in soils 

Soil Sci. Soc. Am. Proc. 32, 35 

Fe203 complex with selenide 

Griffin, R. A. and N. F. Shimp /I 978. Attenuation of pollutants in 
municipal landfill leachate by clay minerals 
EPA/600/2-78/157 (NTIS PB-287140) 
Se, As retention on several clays 

Hammam, A. M. and S. A. Ibrahim /I 980. pH-metric investigation of 

stepwise formation constants of dioxouranium ( IV) , thorium(IV), 

cerium(III), scandium ( II I ) , yttrium(III) and lanthanum ( I I I ) 

chelates with tetrahydroxy-p-benzoquinone 

J, Electrochem. Soc. India 29, 273 

and selenium hbq. Formation, stability constants. 

Husain, S. W. and F. Eivazi /I 975. Thin layer chromatography of 57 

metal ions on an inorganic ion-exchanger in mixed solvent systems 

Chromatographia , 8, 277 

TLC on stannic ar senate- s i 1 ica gel plates. Se, transitions, rare 


Pechkovskii, V. V.; G. F. Pinaev; I. P. Narkevich; S. M. Kirov 

/I 972. Regenerating sorbents 

Patent: USSR SU 354871 

regeneration of zeolites, by dehydrogenation, for selenium oxide 


Rudnev, N. A.; V. N. Pavlova and V. I. Murasheva /I 978. Role of 
the electronic structure of atoms in coprecipitation and sorption 
VINITI 3401-78 

metal hydroxide coprecipitation of Se and Te. Dielectric 
polarizibility . I ^ 

Salter, P. F.; L. L. Ames; J. E. McGarrah /1981. Sorption behavior 
of selected radionuclides on Columbia River Basalts'; 
RHO-BWI-LD-48 (report: Rockwell/Hanf ord ) 
Form not given. Basalts don't retard Se migration. 

Salter, P. F.; L. L. Ames; J. E. McGarrah /1981. Sorption of 

selected radionuclides on secondary minerals associated with the 

Columbia River basalts. 

RHO-BWI-LD-43 (report: Rockwell/Hanf ord ) 

Form not given. Oxidizing conditions, Se not retarded. 


Spoljaric, N. and W. A. Crawford 1^919. Removal of metals from 
laboratory solutions and landfill leachate by greensand filters 
Del. Geol. Surv. , Rep. Invest. 32,1 
Se oxides, pH 3,10. landfill, greensand: 80% glauconite 

Sukharev, Yu. I. and A. I. Volovich /I 980. Inorganic sorbents 

specific for selenium and tellurium 

Khimiya i Tekhnol. Neorgan. Sorentov 1980, 40 

see Ref.Zh., Khim. 1981 :3B1 631. Zr oxide hydrate, phosphate 

Ueshima, H.; S. Shintaro; S. Muranishi; H. Murakami; M. Okamoto 

/1978. Recovery and recycling of selenium values from a spent 

aluminum electrolytic coloring bath 

Patent, Japan Kokai Tokkyo Koho JP 79155945 

At pH<6, adsorbs on weak base resin Daiyaion WA-20 (from 

Mitsubishi Chem.) 

Wilmoth, R. C; J. L. Kennedy; J. R. Hall; C. W. Steuve /I 979 
Removal of trace elements from acid mine drainage 
EPA/600/7-79/101. PB-299194. (NTIS) 
Se, ro, ix, lime, neutralization 

Wilmoth, R, C; T. L. Baugh and D. W. Decker /1979. Removal of 

selected trace elements from acid mine drainage using existing 


Proc. Ind. Waste Conf. 33,886 

fossil fuels, lime neutralization, ix, ro 

Wolff, M. /1978. Selective elimination of heavy metals in waste 
water using ion-exchange resins and adsorbants 
Tribune du Cebedeau 30, 405 
Strong base resins hold Se. 

Yamashige, T.; Y. Ohmoto and S. Yukisato /1978. Ion-exchange 
separation and determination of selenium in ambient particulates 
by heated-quartz cell-atomic absorption spectrophotometry 
Bunseki Kagaku 27, 607 

airborne particulates in HN03 + H202. To Dowex 50Wx8 (acid form), 
eluted with 0.05 M HN03, reduced with Zn under pressure 0.5 kg/cm2 
->K2Se. mass spec. 




Gerhard Klein 

University of California, Berkeley 

Water Thermal and Chemical Technology Center 

47th & Hoffman Blvd., Richmond, California 94804 


After a brief characterization of the ion-exchange process, its potential 
for selenium removal under ideal conditions is discussed. The effect of 
relaxing various premises is examined. On the basis of research results 
obtained at this Center, the order of magnitude of the cost of using ion 
exchange with conventionally available materials is estimated. A novel 
regeneration scheme that offers economic and environmental benefits is 
proposed. The potential profitability of pursuing research on selenium removal 
by ion exchange is examined. 


Although selenium pollution of water is not a widespread problem, it can 
become menacing under certain conditions. An example that has recently 
received considerable publicity is the Kesterson Reservoir in California, 
where the occurrence of selenium levels in the order of one tenth to less than 
one milligram per liter has entailed serious environmental consequences. 

One of the methods considered for removal of selenium from water has been 
ion exchange. The present paper briefly characterizes the ion-exchange process 
and examines its potential for selenium removal. Various factors affecting 
this potential are discussed and a rough estimate is made of the cost of 
achieving selenium removal from Kesterson Reservoir water using conventional 
ion-exchange resins and techniques, and utilizing experimental results 
obtained at this Center under the sponsorship of the Environmental Protection 
Agency. A novel regeneration technique is proposed to reverse the exhaustion 
equilibrium, and thus reduce the amounts of exchanger and regenerant needed. 
Research approaches are examined that could lead to the development of more 
selective exchangers and more economic processes. 



Conventional ion exchangers in use today usually are synthetic plastic 
particles with a diameter in the order of one millimeter. Attached to their 
matrix are functional groups, i.e., electrically charged sites that hold ions 
of the opposite charge ("counterions") by electrostatic attraction. Cation 
exchangers hold positively charged cations; anion exchangers, negatively 
charged anions. For the strong-base anion exchangers of greatest direct 
interest here, the total, equilibrium exchange capacity is of the order of 1.5 
gram equivalents per liter of packed bed and their price, in the order of one 
1984 dollar per liter. 

The bond between a counterion and the functional group is considerably 
weaker than most chemical bonds, so that ions of a different species can 
readily exchange for ions initially attached to the functional groups. This 
process is called "ion exchange." 

Anion exchangers can be of the strong- or weak-base types, according to 
their affinity for hydroxyl ions. Weak-base exchangers have a great affinity 
for such ions, and thus can exchange only in acidic solutions. At the pH of 
Kesterson Reservoir water, for instance, hydroxyl ions would seriously compete 
for exchange sites with any other anion species, and a significant proportion 
of the exchange capacity could be occupied by an anion other than hydroxyl ion 
only if the exchanger had a pronounced selectivity for the former. 

Strong-base exchangers again exist in two types. Type I has a 
considerably greater affinity for chloride ion than Type II, while their 
affinity for sulfate ion, relative to ions other than chloride, is similar. 
Type II resins will thus be of greater interest here. 

In some respects, an ion exchanger resembles an ionic solution. In the 
latter, mobile cations and an equivalent amount of anions move about freely. 
In the ion exchanger, however, the functional groups, and the counterions 
attached to them, are essentially fixed. It is from this property that the 
utility of ion exchangers results. 

The most common process arrangement is that of a fixed bed, i.e., a 
column packed with exchanger, through which a solution flows, typically at 
velocities of 2 to 1 gpm/sq.ft. (5 to 25 meters per hour). Assume that the 
solution fed to the column contains Counterion Species F, and that the 
exchanger is initially saturated with Counterion Species P ("presaturation" 
species). The exchanger near the inlet end will become saturated with F, 
while, at least for some time, the region near the outlet end is still in the 
P-form. In the intermediate zone, or zones, the composition will vary. 

When the bed is nearly filled with the feed-ion species ("exhausted"), 
the latter begins to emerge in the effluent. This is called "breakthrough" and 


marks the time when the bed must be "regenerated" to be ready for the next 
"cycle" of exhaustion and regeneration. 

These concepts make understandable the chief benefits that can be derived 
from ion-exchange operations. First, one ion species can be removed from a 
solution and replaced by another, which has an electric charge of the same 
sign. Thus, a toxic ion species can be replaced by a harmless one. Second, the 
total normality (concentration in ion equivalent per unit volume) of a 
solution can be changed. It would thus be possible, in principle, to remove 
ionic selenium from a dilute solution by an anion exchanger, and to regenerate 
with a concentrated solution, thus removing the selenium from the bulk of the 
fluid and concentrating it into a small volume to facilitate disposal. Viewed 
superficially, ion exchange appears to be an attractive possibility for 
concentrating ionic selenium. As will be shown, numerous factors limit this 
possibility severely. The examination of these factors, however, will point 
the way to research approaches that could lead to reduction of these 

Finally, when a cation exchanger in the hydrogen ion form and an anion 
exchanger in the free-base form are mixed, they are able to remove practically 
all of the electrolytes present in the solution passed through the bed. The 
disadvantage of this extremely effective "deionization" process lies in the 
high cost of the strong-acid and strong-base regenerants required. Since this 
cost is proportional to the normality of the solution to be treated, it would 
be excessive for the treatment of Kesterson water. 


As a basis for making order-of- magnitude estimates, the following data 
(cf. Izbicki, 1984) will be assumed for the Kesterson Wildlife Refuge in 
California. For definiteness, specific values have been selected arbitrarily, 
but any error induced thereby is only commensurate with the approximate and 
tentative nature of the conclusions derived. 

The total volume will be taken to be lo''^ liters, the pH, to be 8.3, and 
the total normality, to be 0.15, the normality of sulfate being 0.1, and that 
of chloride, 0.05. The total selenium concentration ranges approximately 
between 0.1 and below 1 milligrams (2.5 and below 25 microequivalents) 
selenium per liter, implying that the sulfate normality of the water is in the 
order of 10^ times that of selenium. Most of the selenium will be taken to be 
ionic, and the preponderant part of it, to be in the hexavalent (more highly 
oxidized), or selenate, form. 

The tetravalent (less highly oxidized) form is called selenite. Selenate 
ion is divalent and in most respects very similar to sulfate ion, while 
selenite ion, under the conditions prevailing at Kesterson, is monovalent. The 
corresponding biselenite ion is a weak acid, with a pK of 2.46, as compared to 


that of 1.81 of the chemically similar bisulfite ion. The atomic weight of 
selenium is 79. 

Finally, it will be assumed that selenium is to be removed in a thousand- 
day (approximately two-and-a-half-year) period, and that the cost of 
accomplishing this task must not exceed 250 million dollars. 


A systematic experimental and interpretational study of the exchange of 
selenate and selenite has been made by Maneval (1983) and Maneval et al. 
(1985). The salient findings of interest here are related primarily to 
selenate ion because of the much smaller concentration of selenite. 

Only strong-base exchangers were investigated. As expected, since 
selenate and sulfate ions are both divalent, the total solution concentration 
had little effect on their selectivity with respect to a third ion species. 
Also, since biselenate and bisulfate are both fairly strong acids, pH affects 
their uptake only little, and approximately to the same degree. At total 
solution normalities approximately equal to that of Kesterson water, the 
strong-base resins investigated did not discriminate significantly between 
selenate, sulfate, and chloride ions, so that, as an approximation, selenate 
can be considered to obey the tracer principle. 

As a rough approximation, then, the amount (in equivalents) of ion 
exchanger required, if used in a single cycle, would equal the total number of 
anion equivalents in the Reservoir. 

Actually, since the selectivity of the exchanger for all anion species is 
not strictly the same, the system exhibits multicomponent phenomena, and its 
behavior with respect to anion species present in gross concentrations would 
have to be established first. From this, the trace behavior of selenium could 
then be derived. The necessary calculations can be made with the aid of the 
so-called "local-equilibrium theory", which ignores the effect of rate- 
limiting diffusion processes (cf., for example, Helfferich and Klein, 1970). 

A second deviation from ideality is caused by the kinetic effects just 
mentioned. Finally, it is never economically desirable to regenerate a resin 
bed completely, so that the effective capacity often falls significantly below 
that estimated of completely regenerated resin. Also, even to achieve such an 
incomplete degree of regeneration, an excess of regenerant over the 
stoichiometrically equivalent amount is usually required. Minor additional 
causes of inefficiencies are flow irregularities, mixing, etc. General 
methods for the more detailed design of ion-exchange systems have been 
outlined by Vermeulen et al. (1984). 



As a starting point, we shall examine what ion exchange could do if an 
ideally selective exchanger were available, i.e., how much exchanger would be 
required to remove all selenate from the Reservoir, and what the cost involved 
might be. 

Next, recognizing that selenate can approximately be considered as a 
trace in a gross amount of other anions, we can multiply the amount of 
exchanger and cost found for the ideal, completely selective exchanger by the 
ratio of the gross to the trace concentrations to estimate the actual cost for 
a process using commercially available and tested ion-exchange resins. The 
regenerant cost is also estimated, as a multiple of the cost corresponding to 
stoichiometric regeneration. 

The next section deals with the feasibility of regeneration and proposes 
a regeneration scheme that has novel features, but the physical aspects of 
which have received extensive development for an analogous process. 

Finally, factors will be considered that could lead to overall cost 

Selenium Removal by Ideally Selective Exchanger. At an assumed selenate 
normality of 2.5x10"^, the entire Reservoir may be estimated to contain 
2.5x1 0~^x1 0^ ^ = 2.5x10^ equivalents of selenate. This, with a volumetric 
exchange capacity of 1.5 equivalents per liter of packed bed, corresponds to 
2.5x1 0^^/(1 .5n) liters (bulk) of exchanger resin, where n is the number of 
cycles, or to a numerically equal approximate purchase cost of the resin. If 
we furthermore assume f oiu: cycles per day, and a 1 000-day operation, we arrive 
at a resin cost of four dollars and forty six cents to clean up Kesterson 
Reservoir, a sum ridiculously negligible as compared to other, unavoidable 
costs, such as regeneration and pumping. This speculation, though apparently 
idle, nevertheless leads to the conclusion that efforts to increase exchanger 
selectivity for selenate may prove extremely rewarding. 

Removal by Nonselective Exchanger. As has been seen, conventional, 
commercially available, Type-II anion exchangers exhibit little selectivity 
between the ions of interest, so that removal of the bulk of the other anions 
would be required to remove the selenate. However with the regeneration scheme 
proposed further below, regenerant equivalent only to sulfate (plus selenate) 
would be required. 

With the sulfate normality of the Kesterson Reservoir water being about 
0.1 , the resin cost of removing ionic selenium with such exchangers would thus 
equal that just established, times the ratio of the total to the selenate 
normality, i.e., $4.46x40,000 = 0.18 million dollars. If we increase this sum 
by an order of magnitude to allow for such costs as those of pumping. 


equipment, plant construction, waste disposal, and operators, we arrive at two 
million dollars. 

As proposed below, calcium chloride could be a desirable regenerant in 
the present case. The amount of this chemical in the anhydrous state, 
equivalent to the total amount of sulfate in the Reservoir is about 56,000 
tons. Its cost, at the present delivered price of $375 per ton, would be 21 
million dollars. Because of the precipitation reaction which accompanies 
regeneration, only a slight excess of regenerant over the amount corresponding 
to stoichiometric exchange is likely to be needed. A reasonable estimate for 
the total cleanup cost by the ion-exchange scheme proposed might thus be about 
30 million dollars - considerably less than the estimated cost of implementing 
some of the other methods considered. 


Conventional regeneration with sodium chloride might require a several- 
fold stoichiometric excess of regenerant. The following regeneration scheme is 
likely to remedy this difficulty to a large extent, in addition to producing a 
more compact, more easily disposable, and less objectionable waste. Moreover, 
the amount of regenerant would have to be only slightly in excess of that 
corresponding stoichiometrically to sulfate, instead of to all the anions 

Such a scheme suggests itself as a variant of regeneration accompanied by 
precipitation, currently under large-scale development by the California 
Department of Water Resources at Los Bancs. 

The idea, initially proposed by Haugseth and Beitelshees of the U.S. 
Bureau of Reclamation (1974), was conceived independently, put into practice, 
and tested on a bench scale at the University of California, Berkeley (Klein 
et al., 1979), and further developed in cooperation with the California 
Department of Water Resources (State of California, 1978). 

The original application was to the softening of agricultural drainage 
water and was directed toward the regeneration step. The cation exchanger 
primarily in the calcium and magnesium form is regenerated with a waste brine 
of the overall process, so that regenerant chemicals need to be neither 
purchased nor disposed of. 

This brine is composed mainly of a mixture of sodium sulfates and 
chlorides. As sodium displaces calcium from the resin during regeneration, the 
liberated calcium tends to precipitate with the sulfate ion in the brine, thus 
shifting the regeneration equilibrium in the desired direction and providing a 
sink for the calcium ion, which can be disposed of as solid calcium sulfate. 


To avoid the danger of clogging the bed with precipitate, regeneration is 
performed in the upflow direction, at a velocity sufficient to fluidize the 
bed and to provide hydraulic separation of resin and precipitate particles. 

This process has been tested at the bench-scale and pilot-plant-scale 
levels and is presently being implemented at full plant scale at Los Banos. 

The regeneration scheme proposed as a possibility for anion-exchanger 
beds saturated with selenium, sulfate, and chloride would use a fairly 
concentrated solution of calcium chloride as regenerant. The removal of 
sulfate and selenate ions from the exchanger would be brought about not only 
by replacement with chloride ion, but by precipitation of these ions with the 
calcium ion in the regenerant solution, thus entailing a desirable shift in 
equilibrium and providing the major part of the regeneration waste in solid, 
easily disposable, form. Although calcium sulfate is amply available as a 
byproduct of some industries, it does have a market value and if its selenium 
level were tolerable could be used as a soil conditioner in the vicinity of 
the Reservoir. Its disposal would thus be facilitated and a slight economic 
gain could result. 

The regeneration process is thus analogous to the extensively tested 
process being implemented at Los Banos, with the exception that here, there is 
anion instead of cation exchange, and that the regenerant would have to be 
purchased. There is reason to hope that the physical analogy is sufficient to 
make the proposed regeneration process operable with little difficulty. 

An additional major potential advantage of the regeneration procedure 
proposed could be that, if it were desired to subject the pond water to 
reverse osmosis to partially demineralize it, its scaling potential would be 
significantly reduced by the removal of sulfate. 


Even though a first-cut estimate of the cost of removing selenate from 
Kesterson-Reservoir water has shown that a largely proven ion-exchange process 
could be economically attractive, the process not only deserves more detailed 
examination, but appears capable of significant improvement, Gspecially in 
terms of reduced regenerant requirement. Such could be brought about primarily 
by increasing the selectivity of the exchanger for ionic selenium species, 
through modification of the functional groups and the matrix, especially the 
degree of crosslinking, of the resin, or through chemical reactions 
accompanying the ion-exchange process. A change in temperature might also lead 
to a desirable shift in equilibrium. It must be kept in mind, however, that 
simple regeneration becomes the more difficult, the greater the selectivity of 
the exchanger for the ion to be removed. This means that any search for 
heightened selectivity should be paralleled with development of appropriate 
regeneration schemes. 


Clifford and Weber (1982), on the basis of experimental equilibrium 
studies, have developed guidelines as to choosing or synthesizing resins with 
desirable characteristics. This study could be deepened with specific regard 
to selenium removal and emphasis on weak-base resins, synthesis of resins 
satisfying the criteria developed could be attempted, and their performance in 
cyclic operation could be tested under various conditions of practical 
interest. The economic implications of the resulting treatment method should 
be assessed. 

At an early stage in any further development of the regeneration scheme 
proposed, factors such as the effect of ionic activities, solubilities of 
calcium sulfate and selenate under the conditions prevailing, coprecipitation, 
actual regenerability of selenate ion, and the fate of selenate in the overall 
process, should be investigated. 


1 . Because of the preponderance of selenate over selenite ion, the 
reroval of selenate from Kesterson Reservoir water is likely to reduce the 
concentration of ionic selenium in this water to a tolerable level. 

2. At the total solution normality of the water, conventional strong- 
base. Type II anion exchangers exhibit little discrimination between selenate, 
sulfate, and chloride ions. Selenate can therefore be considered as a trace 
ccmponent, and the same fraction of it will be removed as of the gross 
components . 

3. Using state-of-the-art exhaustion and a novel adaptation of a 
regenerant process proven on the pilot-plant scale, effective ion-exchange 
removal of selenate frcm Kesterson Reservoir water in a thousand-day (an about 
two-and-a-half -year) period is estimated to be possible at a total cost of 
about 30 million dollars so that the total probable cost for this process 
appears to compare well with that projected for other types of Kesterson-w^ter 
management . 

4. Since the selenium-containing waste is solid, insoluble, and dilute, 
environmentally acceptable disposal would be practical provided that the 
selenium level is tolerable. 

5. The estimated cost is for an exchanger without selectivity. It would 
be reduced substantially if factors could be identified that would increase 
the specific selectivity for selenate ion, and if an exchanger with such 
selectivity could be synthesized at a reasonable cost. Means of reducing 
selenate to selenite and treatment of the latter could also be investigated. 
If a small fraction of the estimated treatment cost could be allotted to 


research, the chance is considerable that the benefit of the resulting 
findings would more than make up for the expense. 


The valuable discussions with Cal Herrmann, of this Center, who has 
recently completed a draft of an analytical survey of the literature on the 
ion-exchange behavior of selenium, are gratefully acknowledged, as are the 
useful comments of Professor Marshal F. Merriam. 


Clifford, Dennis, and Walter J. Weber, 1982, The Determinants of 
Divalent /Monovalent Selectivity in Anion Exchangers, Reactive Polymers, 1 , 77- 

Helfferich, Friedrich, and Gerhard Klein, 1970, Multicomponent Chromatography, 
Marcel Dekker, New York. 

Izbicki, John A., 1984, Chemical Quality of Water at Fourteen Sites Near 
Kesterson National Wildlife Refuge, Fresno and Merced Counties, California. 
U.S. Department of the Interior Geological Survey Report 48-582. 

Klein, Gerhard, Thomas J. Jarvis, and Theodore Vermeulen, 1979, Fluidized-bed 
Ion Exchange with Precipitation - Principles and Bench-Scale Development, in 
Recent Developments in Separation Science, V, Normal N. Li, editor. CRC Press, 
West Palm Beach, Florida. 

Maneval, James E., 1983, Selenium Removal from Drinking Water by Ion Exchange. 
M.S. Thesis in Chemical Engineering, University of California, Berkeley. (A 
report with the same title and similar content, with coauthors G. Klein and J. 
Sinkovic, is in preparation by the Municipal Environmental Research 
Laboratory, Office of Research and Development, U.S. Environmental Protection 
Agency, Cincinnati, Ohio.) 

State of California Department of Water Resources, 1978, Agricultural Waste 
Water for Power Plant Cooling. Development and Testing of Treatment Process. 

Trussell, R.R., A. Trussell, and P. Kreft, 1980, Selenium Removal from 
Groundwater Using Activated Alumina, U.S. Environmental Protection Agency 
Report EPA-600/2-80-153, Cincinnati, Ohio. 

Vermeulen, Theodore, M. Douglas LeVan, Nevin K. Hiester, and Gerhard Klein, 
1584, Adsorption and Ion Exchange, Section 16 in Perry's Chemical Engineers' 
Handbook, 6th edition, McGraw-Hill, New York.